By: Tzvetan Mikov (tzvetanmi.delete@this.yahoo.com), May 13, 2007 5:23 pm
Room: Moderated Discussions
Linus Torvalds on 5/13/07 wrote:
---------------------------
>[...]
>>So, in practical terms, is there benefit from using
>>64-bit OS on a machine with, lets say, 2-3 GB RAM ?
>
>Absolutely. No question what-so-ever. If you have 2GB
>of RAM on a 64-bit architecture, you can access it all
>easily in the kernel, and none of it is limited to just
>some special use. And a single big process can also use
>it effectively at the same time, so a database can
>actually have it all mapped without having to play
>windowing games in user space either.
Thanks. This was very informative. I guess I didn't actually realize that the HIGHMEM limit is not 4GB but 1GB-128MB (I guess depending on VMSPLIT_xx).
It appears that as far as kernel complexity is concerned, PAE (CONFIG_HIGHMEM64G) is not that much worse than simply having to support >1GB RAM (CONFIG_HIGHMEM4G).
If anybody is interested in some more details, I found this high level overview:
http://kerneltrap.org/node/2450
Mel Gorman's book "Understanding The Linux Virtual Memory Manager" covers it in some more detail. It is available online here:
http://www.phptr.com/content/images/0131453483/downloads/gorman_book.pdf
It seems the complication is already there and isn't ever going away. All driver code that needs to use the CPU to access process memory outside of the current process context needs to use kmap()/kunmap(), even though on a 64-bit system they don't do anything.
regards,
Tzvetan
---------------------------
>[...]
>>So, in practical terms, is there benefit from using
>>64-bit OS on a machine with, lets say, 2-3 GB RAM ?
>
>Absolutely. No question what-so-ever. If you have 2GB
>of RAM on a 64-bit architecture, you can access it all
>easily in the kernel, and none of it is limited to just
>some special use. And a single big process can also use
>it effectively at the same time, so a database can
>actually have it all mapped without having to play
>windowing games in user space either.
Thanks. This was very informative. I guess I didn't actually realize that the HIGHMEM limit is not 4GB but 1GB-128MB (I guess depending on VMSPLIT_xx).
It appears that as far as kernel complexity is concerned, PAE (CONFIG_HIGHMEM64G) is not that much worse than simply having to support >1GB RAM (CONFIG_HIGHMEM4G).
If anybody is interested in some more details, I found this high level overview:
http://kerneltrap.org/node/2450
Mel Gorman's book "Understanding The Linux Virtual Memory Manager" covers it in some more detail. It is available online here:
http://www.phptr.com/content/images/0131453483/downloads/gorman_book.pdf
It seems the complication is already there and isn't ever going away. All driver code that needs to use the CPU to access process memory outside of the current process context needs to use kmap()/kunmap(), even though on a 64-bit system they don't do anything.
regards,
Tzvetan
Topic | Posted By | Date |
---|---|---|
Rock/Tukwila rumors | mas | 2007/05/05 12:59 PM |
Rock/Tukwila rumors | David Kanter | 2007/05/05 02:33 PM |
Rock/Tukwila rumors | Dean Kent | 2007/05/05 03:35 PM |
K8 vs Win64 timeline | anonymous | 2007/05/05 06:19 PM |
Yes, I misremembered... | Dean Kent | 2007/05/05 10:03 PM |
Rock | Daniel Bizó | 2007/05/06 02:34 AM |
Rock | Dean Kent | 2007/05/06 07:11 AM |
Rock/Tukwila rumors | Joe | 2007/05/06 11:24 AM |
Rock/Tukwila rumors | Dean Kent | 2007/05/06 11:49 AM |
Rock/Tukwila rumors | Linus Torvalds | 2007/05/06 12:09 PM |
Rock/Tukwila rumors | anon | 2007/05/07 01:32 AM |
Rock/Tukwila rumors | Rakesh Malik | 2007/05/07 09:36 AM |
Rock/Tukwila rumors | Michael S | 2007/05/07 10:06 AM |
Rock/Tukwila rumors | anon | 2007/05/07 09:48 PM |
Rock/Tukwila rumors | Rakesh Malik | 2007/05/08 06:45 AM |
Rock/Tukwila rumors | anon | 2007/05/08 05:30 PM |
Wow. (nt) | Brannon | 2007/05/08 06:16 PM |
Rock/Tukwila rumors | rwessel | 2007/05/08 09:48 PM |
Rock/Tukwila rumors | JS | 2007/05/08 10:07 PM |
Rock/Tukwila rumors | JS | 2007/05/09 06:44 AM |
Rock/Tukwila rumors | Rakesh Malik | 2007/05/09 05:35 AM |
Much ado about x | Michael S | 2007/05/09 09:39 AM |
Call it x86-64 | Linus Torvalds | 2007/05/09 10:27 AM |
(i)AMD64 | Michael S | 2007/05/09 12:16 PM |
(i)AMD64 | Linus Torvalds | 2007/05/09 12:29 PM |
(i)AMD64 | Groo | 2007/05/09 04:45 PM |
TIFNAA | anonymous | 2007/05/09 05:49 PM |
Inspired by FYR Macedonia? (NT) | Michael S | 2007/05/09 11:21 PM |
More likely... | rwessel | 2007/05/10 12:39 AM |
TIFNAA | Gabriele Svelto | 2007/05/09 11:57 PM |
(i)AMD64 | James | 2007/05/10 02:27 AM |
i86 | Dean Kent | 2007/05/09 12:30 PM |
(i)AMD64 | Max | 2007/05/09 01:28 PM |
wide86? long86? | hobold | 2007/05/10 05:05 AM |
x87 perhaps, it is one more. :) (NT) | Groo | 2007/05/10 05:50 AM |
x86+ | Dean Kent | 2007/05/10 08:44 AM |
Does it really matter? | Doug Siebert | 2007/05/10 09:10 AM |
let's stay with x86-64 for now, please | Marcin Niewiadomski | 2007/05/10 11:50 AM |
let's stay with x86-64 for now, please | Dean Kent | 2007/05/11 06:11 AM |
let's stay with x86-64 for now, please | rwessel | 2007/05/11 02:46 PM |
let's stay with x86-64 for now, please | Dean Kent | 2007/05/11 06:03 PM |
let's stay with x86-64 for now, please | Michael S | 2007/05/12 10:49 AM |
let's stay with x86-64 for now, please | Dean Kent | 2007/05/12 01:05 PM |
let's stay with x86-64 for now, please | Michael S | 2007/05/12 01:25 PM |
let's stay with x86-64 for now, please | Dean Kent | 2007/05/12 03:39 PM |
let's stay with x86-64 for now, please | JasonB | 2007/05/13 07:43 AM |
client consolidation | Michael S | 2007/05/13 08:37 AM |
let's stay with x86-64 for now, please | Tzvetan Mikov | 2007/05/13 03:44 PM |
let's stay with x86-64 for now, please | rwessel | 2007/05/14 02:42 PM |
What's your point? | Doug Siebert | 2007/05/11 02:56 PM |
What's your point? | Linus Torvalds | 2007/05/11 04:15 PM |
What's your point? | Doug Siebert | 2007/05/13 03:11 PM |
What's your point? | Dean Kent | 2007/05/13 07:04 PM |
What's your point? | JasonB | 2007/05/14 02:06 AM |
What's your point? | Dean Kent | 2007/05/14 07:20 AM |
What's your point? | JasonB | 2007/05/14 04:35 PM |
What's your point? | JasonB | 2007/05/14 07:35 PM |
What's your point? | Dean Kent | 2007/05/14 08:12 PM |
What's your point? | Dean Kent | 2007/05/11 06:06 PM |
What's your point? | Stephen H | 2007/05/13 01:55 AM |
Why didn't MS take advantage of PAE? | David W. Hess | 2007/05/13 08:37 AM |
PAE sucks (Why didn't MS take advantage of PAE?) | Linus Torvalds | 2007/05/13 10:20 AM |
PAE sucks (Why didn't MS take advantage of PAE?) | Dean Kent | 2007/05/13 10:49 AM |
PAE sucks (Why didn't MS take advantage of PAE?) | David W. Hess | 2007/05/13 12:37 PM |
> 1 GB RAM on a 32-bit system | Tzvetan Mikov | 2007/05/13 01:44 PM |
> 1 GB RAM on a 32-bit system | S. Rao | 2007/05/13 03:00 PM |
> 1 GB RAM on a 32-bit system | Tzvetan Mikov | 2007/05/13 05:32 PM |
> 1 GB RAM on a 32-bit system | S. Rao | 2007/05/14 12:19 AM |
> 1 GB RAM on a 32-bit system | Linus Torvalds | 2007/05/13 03:46 PM |
> 1 GB RAM on a 32-bit system | Tzvetan Mikov | 2007/05/13 05:23 PM |
> 1 GB RAM on a 32-bit system | JasonB | 2007/05/13 06:37 PM |
Windows manages memory differently | Tzvetan Mikov | 2007/05/13 08:31 PM |
Windows manages memory differently | JasonB | 2007/05/14 01:50 AM |
Windows manages memory differently | Tzvetan Mikov | 2007/05/14 08:56 AM |
Windows manages memory differently | rwessel | 2007/05/14 03:40 PM |
Windows manages memory differently | David W. Hess | 2007/05/14 04:07 PM |
Windows manages memory differently | rwessel | 2007/05/14 04:51 PM |
Windows manages memory differently | Tzvetan Mikov | 2007/05/14 05:40 PM |
Windows manages memory differently | rwessel | 2007/05/14 06:09 PM |
Windows manages memory differently | Howard Chu | 2007/05/14 11:17 AM |
Windows manages memory differently | Jukka Larja | 2007/05/14 11:30 AM |
Windows manages memory differently | Tzvetan Mikov | 2007/05/14 01:54 PM |
Windows manages memory differently | Howard Chu | 2007/05/15 03:35 AM |
Windows manages memory differently | Groo | 2007/05/15 07:34 AM |
Anyone know what OS X (10.4, Intel, desktop) does? | Matt Sayler | 2007/05/15 06:23 AM |
Anyone know what OS X (10.4, Intel, desktop) does? | Wes Felter | 2007/05/15 08:37 AM |
Anyone know what OS X (10.4, Intel, desktop) does? | Anonymous | 2007/05/15 10:49 AM |
Anyone know what OS X (10.4, Intel, desktop) does? | anon2 | 2007/05/15 07:13 PM |
PAE sucks (Why didn't MS take advantage of PAE?) | Paul | 2007/05/13 03:40 PM |
PAE sucks (Why didn't MS take advantage of PAE?) | Peter Arremann | 2007/05/13 05:38 PM |
PAE sucks (Why didn't MS take advantage of PAE?) | Henrik S | 2007/05/14 03:31 AM |
The fragility of your argument | slacker | 2007/05/13 03:56 PM |
The fragility of your argument | nick | 2007/05/13 05:42 PM |
The fragility of your argument | Howard Chu | 2007/05/14 02:52 AM |
The fragility of your argument | Dean Kent | 2007/05/14 09:19 AM |
The fragility of your argument | anon2 | 2007/05/14 08:26 AM |
The fragility of your argument | Tzvetan Mikov | 2007/05/14 09:01 AM |
The fragility of your argument | Dean Kent | 2007/05/14 09:16 AM |
The fragility of your argument | Linus Torvalds | 2007/05/14 11:57 AM |
The fragility of your argument | JasonB | 2007/05/14 04:48 PM |
The fragility of your argument | Dean Kent | 2007/05/14 07:36 PM |
The fragility of your argument | Ricardo B | 2007/05/16 02:40 AM |
The fragility of your argument | Dean Kent | 2007/05/16 03:32 AM |
The fragility of your argument | Ricardo B | 2007/05/16 06:41 AM |
PS | Ricardo B | 2007/05/16 06:50 AM |
The fragility of your argument | Dean Kent | 2007/05/16 09:07 AM |
Modern web browsing | S. Rao | 2007/05/16 09:16 AM |
Aha! | Dean Kent | 2007/05/16 09:27 AM |
Aha! | Dean Kent | 2007/05/16 09:32 AM |
Aha! | S. Rao | 2007/05/16 10:34 AM |
The fragility of your argument | Ricardo B | 2007/05/16 10:00 AM |
The fragility of your argument | Vincent Diepeveen | 2007/05/16 10:10 AM |
The fragility of your argument | Paul | 2007/05/16 03:01 PM |
The fragility of your argument | Vincent Diepeveen | 2007/05/17 03:05 AM |
The fragility of your argument | anon2 | 2007/05/15 01:35 AM |
Splits vs page allocations? | Matt Sayler | 2007/05/15 07:33 AM |
What's your point? | Michael S | 2007/05/13 08:55 AM |
What's your point? | anonymous | 2007/05/13 11:08 AM |
What's your point? | Michael S | 2007/05/13 11:31 AM |
let's stay with x86-64 for now, please | JasonB | 2007/05/13 07:16 AM |
x864 =) (NT) | some1 | 2007/05/15 03:03 AM |
Rock/Tukwila rumors | IntelUser2000 | 2007/05/06 02:27 PM |
Rock/Tukwila rumors | m | 2007/05/13 08:05 AM |
Rock/Tukwila rumors | mas | 2007/05/15 09:40 AM |